Business Statistics

Ikram-E-Khuda
Agenda Items
• Types of hypothesis
• Statistical hypothesis errors
• Z test and t test
• Procedure of hypothesis testing
• One tailed and two tailed tests
• One sample z test and t test
• Independent samples t test
• Practice problems
• Working on SPSS
 Hypothesis Testing
• The main purpose of statistics is to test hypothesis

• A research hypothesis which stands the test of time eventually

becomes a theory
Types of Hypotheses
• Null Hypothesis or Ho
• It is the commonly accepted fact.
• Researchers work to nullify or reject or invalidate it.
• Therefore it is called the null hypothesis.
• It does not mean that the statement is null itself.

• Experimental or Alternate Hypothesis H1 or HA

• the alternate hypothesis will just be the opposite of the null hypothesis.
• For example, the null hypothesis might be “There was no change in the water level this
Spring,” and the alternative hypothesis would be “There was a change in the water level
this Spring.”
 Researchers work to reject, nullify or
disprove the null hypothesis.

Researchers come up with an alternate

hypothesis, one that they think explains
a phenomenon.
Hypothesis Testing
Example
Statistical Hypothesis Errors
 Statistical Hypothesis Errors
Type I and type II errors. ... In statistical hypothesis testing, a type I error is the incorrect rejection of a true null
hypothesis (also known as a "false positive" finding), while a type II error is incorrectly retaining a false null
hypothesis (also known as a "false negative" finding)
Statistical Hypothesis Errors
 Statistical Hypothesis Errors

Type I error probability value is also called false positive or FP rate.

Type II error probability value is also called false negative or FN rate.

The power of any test of statistical significance is defined as the probability that it will reject a false null
hypothesis. Statistical power is inversely related to beta or the probability of making a Type II error. In
short, power = 1 – β.
Statistical Hypothesis Errors
Statistical Hypothesis Errors

TYPE I ERROR: If p value ≥ α

(same as p < β )and rejecting null
hypothesis.

TYPE 2 ERROR: If computed p

value ≤ α (same as p > β) and
accepting null hypothesis.
 Statistical Hypothesis Errors
Type I and Type II errors are inversely related: As one increases, the other decreases. The Type I, or α (alpha),
error rate is usually set in advance by the researcher. The Type II error rate for a given test is harder to know
because it requires estimating the distribution of the alternative hypothesis, which is usually unknown.

Example:
Two drugs are to be compared in a clinical trial for use in treatment of disease X. Drug A is cheaper than Drug
B. Efficacy is measured using a continuous variable, Y, and .H0: μ1=μ2.

Type I error—occurs if the two drugs are truly equally effective, but we conclude that Drug B is better. The
consequence is financial loss.

Type II error—occurs if Drug B is truly more effective, but we fail to reject the null hypothesis and conclude
there is no significant evidence that the two drugs vary in effectiveness. What is the consequence in this case?
Statistical Hypothesis Errors
 Z Test and T Test
• So far we have seen that a hypothesis refers to a supposition (𝐻𝑜 )which is
to be accepted or rejected.

• There are two testing procedures for it

• Parametric test
• makes assumptions of the population distribution from which sample is taken
• Non parametric test
• makes no assumptions of the population distribution from which sample is taken

• In the parametric test there can be two types of tests

• Z-test
• T-test
 Comparison Chart for Z Test and T Test
• This information is particularly for comparing of means but can be generalized for other
parameters’ estimations
ONE SAMPLE TEST OF MEANS
1 Sample Z Tests
and
1 Sample T Tests
 One Sample Test of Means
• A one sample test of means compares the mean of a sample to a pre-
specified value and tests for a deviation from that value.

• For example we might know that the average weight for white babies
in the US is 3,410 grams and wish to compare the average birth
weight of a sample of black babies to this value.
 One Tail and Two Tail Tests
Upper-tailed, Lower-tailed, Two-tailed Tests

The research or alternative hypothesis can take one of three forms. An investigator
might believe that the parameter has increased, decreased or changed. For example, an
investigator might hypothesize:

H1: μ > μ 0 , where μ0 is the comparator or null value (e.g., μ0 =191 in an example about
accepted average weight in men ) and an increase is hypothesized - this type of test is
called an upper-tailed test;

H1: μ < μ0 , where a decrease is hypothesized and this is called a lower-tailed test; or

H1: μ ≠ μ 0, where a difference is hypothesized and this is called a two-tailed test.

The exact form of the research hypothesis depends on the investigator's belief about the
parameter of interest is set up by the investigator before any data are collected.
Rejection Region Approach to Hypothesis Testing
 Important Terms
• Rejection Region: The null hypothesis is rejected if the test statistic lies within this region

• Critical value is the boundary of the rejection region. This could be either 𝑧 ∗ 𝑜𝑟 𝑡 ∗
• The critical region encompasses those values of the test statistic that lead to a rejection of the null hypothesis

• The significance level or α can have values e.g. 0.1, 0.05, 0.01 for 90%, 95% 99% confidence intervals respectively.

• A value of α implies that the null hypothesis is rejected α % of the time when it is in fact true.

• Therefore a precise statistical test requires that α% should be as little as possible

• The p-value is the probability of obtaining the test statistic ( z or t value) given that the null hypothesis is true.
• P values summarize statistical significance
• A small p-value is an indication that the null hypothesis is false.
• To make null hypothesis false, p value should be lesser than α

• P values can be calculated manually but we will use software to find them

All these quantitative metrics are defined before the hypothesis test is performed.
 Important Terms
The following figures illustrate the rejection regions defined by the decision rule for upper-,
lower- and two-tailed Z tests with α=0.05. Notice that the rejection regions are in the upper,
lower and both tails of the curves, respectively. The decision rules are written below each figure.
 Procedure for Hypothesis Testing
The procedure for hypothesis testing is based on the ideas described above.
Specifically, we set up competing hypotheses, select a random sample from the
population of interest and compute summary statistics. We then determine whether
the sample data supports the null or alternative hypotheses. The procedure can be
broken down into the following five steps.

Step 1. Set up hypotheses and select the level of significance α.

H0: Null hypothesis (no change, no difference)
i.e. 𝐻𝑜 : 𝜇 = 𝜇𝑜
H1: Research hypothesis (investigator's belief); α =0.05 or any
i.e. 𝐻1 : 𝜇 ≠ 𝜇𝑜 (two tailed test)
𝐻1 : 𝜇 > 𝜇𝑜 (one tailed and upper one tailed test)
𝐻1 : 𝜇 < 𝜇𝑜 (one tailed and lower one tailed test)

ഥ.
Step 2. Calculate the sample mean ; 𝒙

σ𝒏
𝒊=𝟏 𝒙𝒊
ഥ=
𝒙 𝒏
 Procedure for Hypothesis Testing
• Step 3. Select the appropriate test statistic.
The test statistic is a single number that summarizes the
sample information.
An example of a test statistic is the z statistic which is
used when sample size is greater than 30 and population
standard deviation is known.
It is computed as follows:
ҧ 𝑜
𝑥−𝜇
𝑧=
𝜎/ 𝑛

• When the sample size is small and population standard

deviation is not known , we will use t statistics.
𝑥ҧ − 𝜇𝑜
𝑡=
𝑠/ 𝑛
 Procedure for Hypothesis Testing
• Step 4: Rejecting the null hypothesis
• For One Tailed Test
• Using the rejection region 𝛼, we reject the null hypothesis if z or t is greater than
+ 𝑐𝑟𝑡𝑖𝑡𝑖𝑐𝑎𝑙 𝑣𝑎𝑙𝑢𝑒 or less than −𝑐𝑟𝑖𝑡𝑖𝑐𝑎𝑙 𝑣𝑎𝑙𝑢𝑒.

• For Upper One Tailed Test

• Using the rejection region 𝛼, we reject the null hypothesis if z or t is greater than
+ 𝑐𝑟𝑖𝑡𝑖𝑐𝑎𝑙 𝑣𝑎𝑙𝑢𝑒.

• For Lower One Tailed Test

• Using the rejection region 𝛼, we reject the null hypothesis if z or t is less than − critical
value.
 Procedure for Hypothesis Testing
Step 5. Conclusion.

The final conclusion is made by comparing the test statistic (which is a summary
of the information observed in the sample) to the decision rule. The final
conclusion will be either to reject the null hypothesis (because the sample data
are very unlikely if the null hypothesis is true) or not to reject the null hypothesis
(because the sample data are not very unlikely).

If the null hypothesis is rejected, then an exact significance level is computed to

describe the likelihood of observing the sample data assuming that the null
hypothesis is true. The exact level of significance is called the p-value and it will be
less than the chosen level of significance if we reject H0.
 Procedure for Hypothesis Testing
Method 2
• We can also calculate any percentage of confidence interval (91%, 95%, 99%
or any other) around the mean.

• The general form for a confidence interval around the mean, if 𝜎 is known, is
∗
𝜎
𝑥ഥ ± 𝑧
𝑛

• The general form for a confidence interval around the mean, if 𝜎 is

unknown, is
∗
𝑠
𝑥ഥ ± 𝑡
𝑛

• If the computed values in any of the above equations lie in the rejection
regions defined by the respective confidence intervals than null hypothesis
can be rejected otherwise not.
Summary of Comparison Means Statistical Tests
 References
• http://www.probabilityformula.org/central-limit-theorem.html#
• https://www.geoib.com/confidence-level--margin-of-error.html
• https://onlinecourses.science.psu.edu/stat504/node/19/
• http://www.statisticshowto.com/probability-and-statistics/hypothesis-
testing/
• http://blog.minitab.com/blog/adventures-in-statistics-2/understanding-t-
tests-1-sample-2-sample-and-paired-t-tests
• https://keydifferences.com/difference-between-t-test-and-z-test.html
• https://keydifferences.com/difference-between-t-test-and-z-test.html
• http://www.six-sigma-material.com/Hypothesis-Testing.html